Adaptive Transceivers for Wireless Networks

Northwestern University, Evanston IL

Investigators

Abstract

The PI is investigating methods for jointly adapting transmitters and receivers in a wireless network. The objective is to optimize performance by measuring and exploiting properties of the channel and interference. Parameters adapted at the transmitter include data rate and power, along with transmitted signatures, which can span dimensions across space, time, and frequency. An important part of this effort is to account explicitly for limited feedback from the receiver to the transmitter. Both cellular and non-cellular (e.g., peer-to-peer) wireless networks are being considered. In the latter case, the PI is investigating methods for allocating rate and power across degrees of freedom associated with multiple antennas and links in a multi-hop network. Joint power and signature optimization problems are being studied both with perfect channel knowledge at the transmitter, and with limited feedback. Signature optimization serves to avoid interference while simultaneously exploiting channel conditions. Solutions to problems with perfect channel knowledge provide benchmarks for comparisons with suboptimal transmission schemes, e.g., when channels are unknown. Channel models of interest include multiple access (reverse-link) and forward-link channels, and the uncoordinated multi-input/multi-output channel, which models a peer-to-peer network. Transmitter optimization is being studied in the context of both direct-sequence spread spectrum and multi-carrier signaling. Implications for joint transmitter-receiver adaptation over time-varying channels are being considered. Transmitter designs, which can exploit link diversity in a simple multi-hop channel model, are also being explored.